Accessibility
Language
Authors: Florian Kaiser, Mike Zehner, Dr. Oliver Mayer Date: December 2019
With the finite nature of fossil fuels comes the need to switch to renewables. Unfortunately, those with the greatest potential are highly volatile in this regard. Photovoltaics is an extreme case in point. It generates energy only during the day and also has strong fluctuations. One reason for this is the clouds. In this article, the influence of cloud movement on increases in irradiation will be investigated.
To understand the influence of clouds on irradiance, first of all the stationary case LibRadtran Renewable Energy Simulation will be reviewed. Here, a cloud is assumed that does not change and does not move. The variable quantities are thus only the cloud field and the zenith angle. After understanding the simplified case, the clouds can be moved (dynamic case). Hereby one gains two more variables. On the one hand, the duration of the movement can be determined arbitrarily. Second, one can change the cloud velocity.
Such a model is simulated in a ray-tracing program and insights are gained for the irradiance on a PV system.
For a systematic investigation, real measured clouds were used. The cloud cover was varied from 10% to 100% in 10% steps and the zenith angle from 0º to 70º in 10º steps. A period of 300s and 3000s was chosen for the moving clouds. The velocity of the motion was 10 m/s.
From the simulations some conclusions can be drawn:
- For the direct irradiance an almost binary relation with clouds is valid. Either the sun is visible and the cloud has no influence, or the field is shadowed and the direct irradiance drops to near zero. This is true regardless of zenith angle and degree of occultation.
- The behavior of diffuse radiation in interaction with cloud fields is crucial. It turns out that around the shadow edges of a cloud significantly increased values are visible. If the cloud is small enough, the effect can be superimposed in the center of the shadow and maximum results can be obtained there. If the shadow becomes too large, however, the diffuse radiation in the center decreases and at some point even falls below the clear sky case. Not only the horizontal coverage counts for this. The vertical expansion of a cloud can also trigger this effect. If the sky is completely covered, it can become dark during the day. This can be observed well in thunderstorms.
- The zenith angle plays an important role. On the one hand, the diffuse radiation decreases as the angle increases and on the other hand, a low sun increases the cast shadows. This artificially increases the degree of occultation. Combined, these two observations lead to excess irradiance around the shadow of a cloud. The larger the degree of occultation, the higher the global radiation can increase.
An overview is given in Table 12. At 80% coverage, the largest irradiation excess heights are determined. These can reach values up to 407 W/m2. Even at 90%, excess heights above 370 W/m2 are still possible. If the cloud cover becomes even denser, however, these extremes drop again because the shading becomes too great. If there are only single holes left, the values of the maximum global irradiance drop to those of about 30%. Since the zenith angle influences the maximum irradiance, the irradiance excesses decrease with increasing angle. Between 0º to about 20º there is still no big difference. Thus, significant excesses can be measured even in Germany at midday, where the zenith angle is around 25º.
As a further finding, it can be said that much depends on the shape of the overcast cloud. By chance, a cloud field can consistently increase the global radiation or significantly reduce it. This behavior is thereby only conditionally dependent on the considered period and the zenith angle.
The effect usually lasts only a few minutes, but is relevant for the power provision and especially the prediction of the generated power. This is because previous systems calculate with very high inertia and thus set values for the next hour much too low before the radiation overshoot and too high after the overshoot. This leads to fluctuations in the operation of the network.
In order to verify the simulations, measurements were made. This involved observing the sky with a fisheye camera and measuring the power at a PV generator.
Deviations from clear sky days mostly come from shading of clouds. At the edges, however, the light is refracted. Thus it happens that areas without shadows suddenly show a larger global radiation than would be possible on a sunny day. The following measured curves show this clearly.
The global horizontal irradiation is the blue line. You can see a strong fluctuation. Also, the maximum value is well above that of a sunny day. The simulated irradiances from the sun are shown by the red curve. The maximum is about 900 W/m2. The measured data, however, come to over 1,200 W/m². This is due to the increased diffuse radiation, shown in green in the graph. As a result, there is a power supply that is highly dynamic, especially under overcast skies, and must be absorbed by the grid or storage.